DAPHNE: An Open and Extensible System Infrastructure for Integrated Data Analysis Pipelines

Patrick Damme, Marius Birkenbach, Constantinos Bitsakos, Matthias Boehm, Philippe Bonnet, Florina Ciorba, Mark Dokter, Pawl Dowgiallo, Ahmed Eleliemy, Christian Faerber, Georgios Goumas, Dirk Habich, Niclas Hedam, Marlies Hofer, Wenjun Huang, Kevin Innerebner, Vasileios Karakostas, Roman Kern, Tomaž Kosar, Alexander KrauseDaniel Krems, Andreas Laber, Wolfgang Lehner, Eric Mier, Marcus Paradies, Bernhard Peischl, Gabrielle Poerwawinata, Stratos Psomadakis, Tilmann Rabl, Piotr Ratuszniak, Pedro Silva, Nikolai Skuppin, Andreas Starzacher, Benjamin Steinwender, Ilin Tolovski, Pinar Tözün, Wojciech Ulatowski, Yuanyuan Wang, Izajasz Wrosz, Aleš Zamuda, Ce Zhang, Xiao Xiang Zhu

Publikation: Konference artikel i Proceeding eller bog/rapport kapitelKonferencebidrag i proceedingsForskningpeer review

Abstrakt

Integrated data analysis (IDA) pipelines---that combine data management (DM) and query processing, high-performance computing (HPC), and machine learning (ML) training and scoring---become increasingly common in practice. Interestingly, systems of these areas share many compilation and runtime techniques, and the used---increasingly heterogeneous---hardware infrastructure converges as well. Yet, the programming paradigms, cluster resource management, data formats and representations, as well as execution strategies differ substantially. DAPHNE is an open and extensible system infrastructure for such IDA pipelines, including language abstractions, compilation and runtime techniques, multi-level scheduling, hardware (HW) accelerators, and computational storage for increasing productivity and eliminating unnecessary overheads. In this paper, we make a case for IDA pipelines, describe the overall DAPHNE system architecture, its key components, and the design of a vectorized execution engine for computational storage, HW accelerators, as well as local and distributed operations. Preliminary experiments that compare DAPHNE with MonetDB, Pandas, DuckDB, and TensorFlow show promising results.
OriginalsprogEngelsk
TitelConference on Innovative Data Systems Research
UdgivelsesstedSanta Cruz, California, USA
Publikationsdato9 jan. 2022
StatusUdgivet - 9 jan. 2022

Citationsformater